Moderate Safety Question 122 of 223

What is training-data leakage in LLM products?

GenAI / LLM · Speak this in 60–90 seconds · Faridabad & Delhi NCR

PICTURE THIS: DATA SPLIT

Train 70%Val 15%Test 15%

Fit on train, tune on val, report on test once.

Simple meaning

The model may regurgitate copyrighted or private strings it saw in pretraining or fine-tuning.

WHY — Tokens instead of words?

LLMs use tokens (not full words) because it helps them:

Safety questions separate people

who only read docs from people who shipped.

Keep it short, concrete,

and tied to GenAI / LLM work.

Stable token IDs

Each piece maps to a number the network can learn.

Fits the model

Fixed pieces are what transformers expect as input.

STEPS — What happens step by step?

Before you speak the answer, walk the interviewer through these steps:

1
The model may regurgitate
copyrighted or private strings it saw in pretraining or fine-tuning.
2
Users can also trick
it into repeating other tenants' prompts if logs leak into training.
3
Isolation, data minimization, and
provider terms matter.
4
Give an example
One tiny concrete case you can say aloud.
5
Common mistake
What juniors usually get wrong.
6
Close
When you pick this over the alternative.

EXAMPLE — See it in action

Here's a short line you can speak, broken into clear beats:

Say this line

“Users can also trick it into repeating other tenants' prompts if logs leak into ”

Break into beats

Userscanalsotrickitinto

Speaking order

2987408337471632900

Note: Adapt this scaffold to your own project — keep it under 60–90 seconds.

Key takeaway

The model may regurgitate copyrighted or private strings it saw in pretraining or fine-tuning. Users can also trick it into repeating other tenants' prompts if logs leak into training.